<<

MINERALOGICAL SOCIETY OF AMERICA, SPECIAL PAPER 1, 1963 INTERNATIONALMINERALOGICAL ASSOCIATION,PAPERS, THIRD GENERAL MEETING

NORITIC BODIES IN THE SIERRA NEVADA BATHOLITH

ALDEN A. LOOMISl Department of Geology, Stanford University, Stanford, California

ABSTRACT A group of small noritic plutons were intruded prior to the immediately surrounding granitic rocks in a part of the composite Sierra Nevada batholith near Lake Tahoe. Iron was more strongly concentrated in the late fluids of individual bodies than during the intrusive sequence as a whole. are more ferrous in rocks late in the sequence, although most of the iron is in late magnetite which replaces pyroxenes. Both Willow Lake type and normal cumulative layering are present. Cumulative layering is rare; Willi ow Lake layering is common and was formed early in individual bodies. Willow Lake layers require a compositional uniqueness providing high ionic mobility to explain observed relations. The Sierran norites differ from those in large stratiform plutons in that (1) the average bulk composition is neritic anorthosite in which typical rocks contain over 20% AbO" and (2) differentiation of both orthopyroxene and produced a smooth progressive sequence of compositions. A plot of modal An vs. En for all the rocks in the sequence from early Willow Lake-type layers to late no rite dikes defines a smooth non-linear trend from Anss-En76 to An,,-En54. TLe ratio An/ Ab decreased faster than En/Of until the assemblage Anso-En65 was reached and En/Of began to decrease more rapidly. Similar plots for large stratiform bodies show too much scatter to define single curves. Temperature gradients of 40°-50° C. within the magma chambers and variable convection patterns could account for the observed variations.

INTRODUCTION hornblende. The largest quartz diorite body contains A group of consanguineous intrusions crops out local Willow Lake type combed layers as rims aroun \ around Eagle Lake, just southwest of Emerald Bay, inclusions and as thin sheets on internal intrusive Lake Tahoe in the fifteen-minute Fallen Leaf Lake contacts. The present volume of the earlier noritic quadrangle, California. The rock types in the se- rocks is only about 10 per cent of that of the later quence, from oldest to youngest, are (1) noritic quartz diorites. anorthosite and leuconorite, (2) hypersthene diorite, (3) hornblende quartz diorite and (4) hornblende- COMPOSITION AND TEXTURES biotite quartz diorite with minor granodiorite. The The rocks of the noritic sequence are much more two oldest groups of rocks will be called the Eagle feldspathic than the more common early and Lake sequence; their generalized relations with diorites of this part of the Sierra Nevada (Loomis, other rocks is shown in Fig. 1. The noritic anortho- 1961). Compton (1961) and Taubeneck and Polder- site-hypersthene diorite sequence comprises many vaart (1960) have reported similar noritic rocks from small bodies with complex intrusive relations among the northern Sierra Nevada and the Wallowa batho- one another. The bodies are not separated in Fig. 1. liths, respectively, and rocks of this composition may The term noritic anorthosite is used for rocks with prove to be more common than originally though t. color index more than 10 and less than 22t (Budding- The more leucocratic norites from the San Marcos ton, 1939, p. 19), and plagioclase more calcic than of the Southern California batholith (Larsen, An50. Those rocks with plagioclase more sodic than 1948) are also quite similar. All have similar textures An50 are called hypersthene diorites. The entire and appear to be early relative to their immediate series is of interest because of its unusual composition surroundings. Modes, chemical analyses, and norms considering its position in the composite Sierra of two hypersthene diorites are given in Table 1. Nevada batholith, its layered structures, and its con- Each body in the Eagle Lake sequence is struc- sistent trend of mineral compositions with decreasing turally an individual intrusion with foliation which age. generally parallels contacts and truncates structures The quartz diorite and minor granodiorite bodies in older adjacent bodies. Combed Willow Lake-type listed as numbers 3 and 4 above intrude the noritic layering (Taubeneck and Poldervaart, 1960) lies rocks. They are shown as younger intrusive rocks in along many intrusive contacts within the complex, Fig. 1. Many of the quartz diorite bodies have hypers- and mayor may not be broken and deformed by thene grains rimmed with olive or brownish olive later magmatic movements. Cumulative layering

1 Present address: California Institute of Technology, Jet has been found in large blocky inclusions in younger Propulsion Laboratory, Pasadena, California. noritic rocks. The foliate and lineate fabrics of indi- 62 SIERRA NEVADAN 63

MORAINE

NORITIC SEQUENCE

I~ \ ~ x I OLDER INTRUSIVE ROCKS

CONTACT, DASHED WHERE ~// APPROXIMATE »> FAULT

ROAD ~

( - TRAIL "- _/

STATE OF CALIFORNIA N SHOWING LOCATION OF o 5000 10000 FALLEN LEAF LAKE -----. FEET 15' QUADRANGLE r GEOLOGIC MAP SHOWING LOCATION OF EAGLE LAKE NORITIC SEQUENCE

FIG. 1. Generalized geologic map of the northeast part of the Fallen Leaf Lake 15-minute quadrangle.

TABLE 1. MODES, ANALYSES AND NORMS FOR Two SPECIMENS FROM THE EAGLE LAKE SEQUENCE

Modes, Vol. % Analyses, WI. % Norms, WI. % 593 600 593 600 593 600

plagioclase 81.7 75.8 Si02 52.2 51.3 apatite 0.3 0.7 orthoclase 0.7 0.1 AJ,03 21.7 20.2 ilmenite 0.6 2.0 quartz none 0.4 Ti02 0.3 1.1 magnetite 4.9 5.6 hypersthene 9.5 8.4 Fe203 3.3 3.9 orthoclase 3.3 2.1 2.7 1.7 FeO 5.2 4.3 plagioclase 77 .1 70.8 hornblende 1.0 2.5 MnO nd 0.12 diopside 2.3 5.3 biotite 0.3 1.1 MgO 3.8 3.6 hypersthene 11.1 12.7 magnetite 4.0 4.9 CaO 8.8 8.6 quartz 1.2 0.8 apatite 0.1 0.2 Na20 4.0 4.35 calcite, chlorite none 5.0 K20 0.5 0.35 normative An 50 47

H2O+ nd 1.4 normative En 66 70 modal An 45 45 p,Os 0.1 0.30 modal En 60 68 CO, nd 0.54 density 2.88 2.84 Total 99.9 100.06

Analysis 593 by Albert G. Loomis. Analysis 600 by E. L. P. Mercy. 64 ALDEN A. LOOMIS

vidual plutons are not directly related to either type of layering. Most of the plagioclase and orthopyroxene in the plutons crystallized before development of foliation. The bulk of the augite present crystallized after the bodies had been emplaced and the fabrics had been developed. Subhedral and euhedral plagioclase tablets define a flow foliation (exclusive of the layered structures) and c-axes of subhedral or hypersthene grains define a lineation in the foliation plane. Augite occurs both as discrete crystals or as jackets surrounding the orthopyroxenes. Much augite is poikilitic, enclosing several and plagio- clase grains. Olive-green to brownish-olive horn- FIG. 3. Irregular mode of amphibolization in hypersthene blende has jackted and partially replaced the pyrox- diorite near contact with another hypersthene diorite at lower left. enes and plagioclase. Hornblende rims are commonly continuous around several grains which are aligned in magnetite is late; it replaces and is interstitial to the fabric; therefore, most, if not all, of the horn- earlier phases. The small colorless grains in the blende crystallized after the internal movements of are apatite. The amount of magnetite varies very the bodies had ceased. Amounts of hornblende vary; little within the complex, generally forming 4 to 5 amphibolization is described below. Magnetite volume per cent of the rocks. The amount of horn- crystallized subsequently or nearly contemporane- blende is quite variable, however, as is the presence of ously with hornblende. It replaces pyroxenes and deuteric alteration. The rocks are amphibolized par- hornblende and is interstitial to plagioclase, py- ticularly along contacts and fractures, as in Fig. 3. roxenes, and hornblende. Late deuteric alterations Figure 4 is a cumulative mode diagram of 14 speci- are local. Common include chlorite, actino- mens in varying states of alteration. The extent of lite, biotite, and calcite. actual replacement of plagioclase by hornblende, not Figure 2 illustrates a common texture in the noritic only the reaction of pyroxene and plagioclase to anorthosite. Single hornblende grains surround hornblende, is illustrated by the changes from the pyroxenes and replace plagioclase. The bulk of the fresh rock No. 593 to the amphibolized No. 622. Al- though the volume percentage of hornblende in-

.,c: .,o Q. W ::;; ::J _j o >

60~~VO~'~~6~2~7~7~7~6~5~4'A'6~2~3~6~2~2~6~22'-~'6~2~2-L_'6okoO-~~6~5~6~6"'5~4'"5~94~593

SPECIMEN NUMBER

FIG. 2. Noritic anorthosite specimen 622. Olive-green hornblende FIG. 4. Cumulative modes of 14 noritic anorthosites and hyper- rims pyroxenes. Magnetite, commonly with apatite inclusions, is sthene diorites which had very similar modes prior to amphiboliza- late, fills voids, and replaces pre-existing grains. tion and deuteric uralite, quartz and biotite. SIERRA NEVADAN ANORTHOSITES 65 creases from a minimum of zero per cent to a maxi- mum of 13 per cent, the volume of the pyroxene shows a complementary decrease of 4 per cent or less. The volume now occupied by green hornblende was originally occupied mostly by plagioclase. The pyroxenes remaining after the amphibolization and introduction of late magnetite were uralitized locally, and some deuteric biotite and quartz formed. These alterations took place mainly near contacts with older plutons and where the rocks were intruded by the surrounding quartz diorites. Iron was more strongly concentrated during the crystallization of anyone body than during the entire intrusive sequence. The late magnetite in each body FIG. 6. Combed layers along intrusive contact (covered at left). is more important than the effect of the iron increase Hammer is in foliation plane of younger rock which contains due to changing orthopyroxene composition during layers. the whole sequence. Somewhat more than half of the bulk iron in the rocks exists in the late magnetite. As layering is very subtle and is only expressed by each individual body crystallized, the iron was changes in color index over thicknesses on the order of strongly concentrated in the late fluid. The entire 1 cm. Cross-bedding indicating current deposition is sequence comprises many small intrusions, and the present locally and in anyone outcrop the tops are 0 Fe/Mg ratio increased slowly throughout the se- consistent. Figure 5 shows bedding dipping 20 to the quence as will be described below. But iron enrich- left truncated at the top by horizontal layers. The ment was much more extreme within each small body cumulative layers contain the most magnesian ortho- of magma after it separated from the main mass and pyroxene and most calcic plagioclase in the sequence. crys tallized independently. No has been found. Augite is much less abun- dant than the bronzite, as is the case in normal rocks LAYERED STRUCTURES later in the sequence. The Willow Lake-type layers, The noritic rocks contain both normal cumulative however, have much more clinopyroxene than ortho- layering which was formed early in the sequence, and pyroxene. Willow Lake-type layering which formed early in The Willow Lake layers (Fig. 6) lie along intrusive each individual body. The cumulative layers are now contacts and form rims around inclusions which may seen largely as inclusions in younger norites. They be of Willow Lake layers themselves. The crystals have very good foliation defined mainly by plagi- stand normal or at high angles to the plane of the oclase tabular on (010). Rhythmic compositional layering. The plane of the layering is commonly vertical. The thickness of most individual layers is 0.1 to 2 inches. Groups consist of 5 to 20 individual layers and average about one foot thick, although some are 5 feet thick. When followed along strike, a group of layers may diverge smoothly, include a lens of normal rock, and reconverge. A given group normally contains the same number of individuals no matter how much its thickness varies; the individuals thicken or thin together as the group thickens or thins. The variation in thickness of a group is the re- sult of the crystals in the layers having varied in length in the same way in all of the layers. The Willow Lake layers are distinguished from the better known harrisitic structures in that the latter are up- ward growth from previously deposited cumulate :1 6 layers. Descriptions of harrisitic rocks indicate that II "I "Ion II (H G ~ 1.. a v " a nucleation takes place on accumulated grains; the FIG. 5. Hand specimen with cross-laminated cumulative layers. large crystals grow upward but commonly become 66 ALDEN A. LOOMIS

nants of pyroxenes and the pyroxenic layers immedi- ately above and below are quite free of hornblende. The texture of the hornblende is unlike the textures of augite in these layers. The hornblende crystals have a grea ter length to width ra tio and tend to form clus- ters of branching crystals. The elongated are commonly curved. Both species are commonly re- peatedly twinned on (100). Augite is much more abundant than bronzite and forms feathery crystals whereas bronzite does not. The preferential occur- rence of augite, which is subordinate to ortho- pyroxene throughout the rest of the sequence, sug- gests that special conditions obtained during crystal- lization of the combed layers. Presumably, crystals which could nucleate and grow quickly were favored. The combed texture suggests, but does not prove, rapid crystallization on solid surfaces. The sym- pathetic thinning or thickening of several layers (shortening or elongation of crystals) indicates that several layers could grow while some local environ- mental condition on the order of several feet tended to promote or retard growth. The occurrence of pri- mary hornblende layers indicates that water pressure became high at times. The occurrence of pyroxenes in layers adjacent to hornblende layers and small amounts in hornblende layers themselves suggests that water pressure was maintained close to the amphibole-pyroxene phase boundary by crystalliza- FIG. 7. Branching bytownite crystals from Willow Lake tion of either phase. layer. Note discontinuous, bladed albite twins. Taubeneck and Poldervaart (1960) have at- tempted to explain combed layering as the result of highly poikilitic and reach sizes of several inches to a undercooling. The main objection to this hypothesis foot (Brown, 1956). is that if undercooling is the sole explanation, why do The plagioclase, augite and hornblende crystals in such structures not occur in other types of rocks, or the Willow Lake layers are long and feathery. Sheaves in those of equivalent composition that show some of crystals are commonly plumose above discrete evidence of chilling against contacts? Combed struc- centers at the base of the layer. Longer branching tures are restricted in the Fallen Leaf Lake quad- crystals truncate shorter ones as they diverge from the base of the layer. Figure 7 shows a common habit of plagioclase. Albite and Carlsbad twinning pre- dominate; many albite twins are discontinuous along the length of the crystal and in three dimensions are curving blades up to 10 mm long. Some plagioclase grains are zoned primarily along their length, and a wave of progressive extinction passes along the bladed albite lamallae on rotation of the stage. The primary mineralogy of the layers is plagio- clase; plagioclase-a ugi te; plagioclase-augi te-hyper- sthene; and brown hornblende-plagioclase. The brown hornblende is primary in some layers. Figure 8 is a photomicrograph of primary brown hornblende and plagioclase with minor augite and magnetite. The FIG. 8. Primary brown hornblende forming Willow Lake hornblende does not contain partially digested rem- layer. Pyroxenes in adjacent layers are not amphibolized. SIERRA NEVADAN ANORTHOSITES 67

90,------,------,------,------,------, rangle to the noritic rocks and the adjacent quartz diorites which are derivative from them. They do not occur in the older gabbro and diorite bodies which were intruded earlier at this level into presumably cooler, wetter surroundings. Neither do they occur in the quartz diorite, granodiorite, and other bodies composing the rest of the batholith. Furthermore, the occurrence of combed layers at interpluton contacts and around inclusions shows ..... / that the cause of the combed textures could not have i':! beenTocal, but must have affected entire bodies if)15 60 / / ----r----t--&---L _.. nearly simultaneously. Thermal transfer to a heat / sink by convection would seem to be too inefficient a I process to require crystallization of the combed layers rather than normal crystal growth, especially if the concentration of water was sometimes high enough to cause precipitation of primary hornblende. Relatively rapid cooling in a small body with a high surface to volume ratio should be expected; the com- , mol % bination of a normally quick cooling rate and a com- FIG. 10. Compositions of coexisting minerals from positional peculiarity producing high ionic mobility literature for large stratiform bodies. might produce the effect. If water pressure was close to load pressure, sudden lowering of water pressure Two of the points with plagioclase more calcic than might produce rapid isothermal crystallization. Anso are from Willow Lake layers and two are from Opening of fractures in wall rocks during rapid em- large inclusions of cumulative layers. Several of those placement of wet magmas from a lower level is the with plagioclase between An53and An70are also from type of structural control envisioned. An attempt to combed layers. The rest are from normal plutonic produce similar combed textures and their variations rocks and late norite dikes intruded into individual experimentally will begin shortly. noritic anorthosite bodies. The points define a main trend and a (dashed) subsidiary trend. The mineral MINERAL COMPOSITION SEQUENCES compositions were determined optically using optic Figure 9 is a plot of modal An and En mol per- angle for the orthopyroxenes and Rittmann and centages in coexisting plagioclase and orthopyroxene Turner methods with a universal stage for the plagi- from 25 specimens from the Eagle Lake sequence. oclase. N either the plagioclase nor the pyroxene is zoned enough to affect the results. The ratio An/ Ab decreased faster than En/Of until the point An50-En65was reached and En/Of began to EAGLE LAKE SEQUENCE decrease more rapidly. Similar plots for most larger stratiform bodies show much more scatter. Figure 80 10 is a plot for Stillwater and Bushveld with the ;f- Eagle Lake trend from Fig. 9 added for comparison. 0 E Stillwater rocks show little progression; Bushveld w- ...... 70 shows a good deal of scatter. Both curves are drawn i= .....« for coexisting early cumulative phases that are zoned if) z w slightly, if at all. Figure 11 shows plots for Skaergaard, Rhum and 60 the . Modal compositions were used where possible, but normative points are included as well as olivine compositions where orthopyroxene data are lacking. Skaergaard nicely illustrates a ANORTHITE, mol % broad variation that diverges more and more with more highly differentiated rocks. The increased diver- FIG. 9. Modal compositions of coexisting plagioclase and ortho- pyroxene from rocks of the Eagle Lake sequence. gence from a single line of points, expressed as a 68 ALDEN A. LOOMIS

90 which would be caused by temperature gradients of xx x 40°-50° C. as estimated from the experimental phase x 80 diagrams. fj. fj. In the Eagle Lake rocks, minerals which are found ,_- - --- x together crystallized and remained together during fj. - 8_ 70 -- )/ 0 emplacement. The larger stratiform bodies cooled ~ / more slowly and contained temperature and com- 0 0 "0 / t? E 60 0 position gradients. In a large body, convection cur- / W ffio 0 rents probably would sweep out different parts of the f- I a:: I~fj. magma at different times, depositing separate layers w Q f- 50 If) which did not crystallize in the same place in the a:: fj. 0 LL chamber. Temperature and consequent composition a:: 40 Q gradients would be produced locally during relative w 0 quiescence of some parts of the chamber with respect f- i= to others. As deposition proceeds and the height

FIG. 11. Compositions of coexisting minerals from ACKNOWLEDGMENTS literature for large stratiform bodies. The work reported was done as part of a doctorate project at Stanford University during the years broadening of the envelope containing the points at 1958-1961. Thanks are due Professor Robert R. the low-temperature end, is to be expected. The Compton of Stanford University for many helpful dis- slopes of the experimentally determined solidus sur- cussions both in the field and laboratory and to faces in systems containing plagioclase, ortho- Dr. Robert C. Speed, of Stanford and the Jet Pro- pyroxene and olivine decrease for the lower temper- pulsion Laboratory, for advice and critical reading of ature phases. Therefore, a given temperature change the manuscript. The writer's expenses were defrayed in a magma should produce progressively a greater by a National Science Foundation fellowship during difference in crystal composition as differentiation 1959-1961. The Shell Fund for Fundamental Re- progresses. The amount of compositional variation in search at Stanford University paid for some thin sec- the Skaergaard assemblages is equivalent to that tions and the analyses.

REFERENCES

BROWN, G. M. (1956) The layered ultrabasic rocks of Rhum, --- (1960) Stillwater igneous complex, Montana. Geol. Soc. Inner Hebrides. Roy Soc. London, Phil. Trans. 240,1-53. Am. Mem. 80. --- (1957) Pyroxenes from the early and middle stages of LARSEN, E. S. (1948) Batholith of southern California. Geol. Soc. fractionation of the , East Greenland. Am. Mem. 29. Mineral Mag. 31, 511-543. LOMBAARD,B. V. (1934) On the differentiation and relationships of BUDDINGTON, A. F. (1939) Adirondack igneous rocks and their the rocks of the Bushveld Complex. Geol. Soc. South Africa 37, 5-52. metamorphism. Geol.Soc. Am. Mem. 7. Trans. LOOMIS, A. A. (1961) Petrology of the Fallen Leaf Lake area, COMPTON, R. R. (1961) Peridotite-gabbro-granodiorite pluton in California. Ph.D. Thesis, Stanford Univ. the Sierra Nevada, California; (abs.) Geol.Soc. Am., Cordilleran TAUBENECK, W. H. AND A. POLDERVAART(1960) Geology of the Section. Elkhorn mountains, northeastern Oregon. Part 2. Willow Lake HALL, A. L. (1932) The of the central intrusion. Bull. Geol.Soc. Am. 71,1295-1322. Transvall. Geol. Surv. South Africa Mem. 28. WAGER, L. R. AND W. A. DEER (1939) Geological investigations HESS, H. H. (1950) Vertical mineral variation in the Great Dyke in E. Greenland. Pt. III, Petrology of the Skaergaard intrusion, of Southern Rhodesia. Geol.Soc. South Africa Trans. 53, 159-166. Kangerdluzssuaq, East Greenland, Medd. Grenland, 105 (4).